IMPLEMENTASI ALGORITMA DIFFERENTIAL EVOLUTION UNTUK PENCARIAN SOLUSI OPTIMAL GLOBAL BEBERAPA FUNGSI DIFFERENTIABLE DAN NON-DIFFERENTIABLE

    Derry Romeo, - (2023) IMPLEMENTASI ALGORITMA DIFFERENTIAL EVOLUTION UNTUK PENCARIAN SOLUSI OPTIMAL GLOBAL BEBERAPA FUNGSI DIFFERENTIABLE DAN NON-DIFFERENTIABLE. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Penelitian ini bertujuan untuk mengimplementasikan algoritma Differential Evolution (DE) dalam pencarian solusi optimal global dari fungsi differentiable dan non-differentiable. Fungsi differentiable yang diteliti terdiri dari fungsi Ackley 1, Beale, Chichinadze, Giunta, Keane, Mishra 6, Shekel 7, dan Zakharov. Sedangkan fungsi non-differentiable yang diteliti terdiri dari fungsi Bartels Conn, Bukin 4, Bukin 6, Price 1, Schwefel 2.21, Step 3, Stepint, dan Xin-She Yang. Algoritma DE bekerja menggunakan beberapa parameter, yaitu jumlah variabel atau dimensi fungsi (N), banyaknya populasi (NP), nilai faktor skala mutasi (F), nilai probabilitas rekombinasi (CR), dan banyaknya iterasi (G). Tahapan dalam algoritma DE adalah insialisasi, mutasi, rekombinasi, dan seleksi. Hasil implementasi menunjukkan bahwa algoritma DE mampu memberikan solusi optimal global untuk hamper semua fungsi, termasuk untuk fungsi yang memiliki banyak titik global minimum. Analisis sensitivitas parameter DE menunjukkan bahwa pemilihan parameter F, CR, NP, dan G sangat berpengaruh terhadap nilai optimal global fungsi dan waktu komputasi.

    This study aims to implement the Differential Evolution (DE) algorithm in finding the globally optimal solution for differentiable and non-differentiable functions. Differentiable functions consist of Ackley 1, Beale, Chichinadze, Giunta, Keane, Mishra 6, Shekel 7, and Zakharov. Non-differentiable functions consist of Bartels Conn, Bukin 4, Bukin 6, Price 1, Schwefel 2.21, Step 3, Stepint, and Xin-She Yang. The DE algorithm uses several parameters, like variables function (N), the number of populations (NP), scale factor (F), the cross-over rate (CR), and the number of iterations (G). DE algorithm consists of several processes: initialization, mutation, cross-over, and selection. The computational results show that the DE algorithm works well in finding globally optimal solutions, especially for functions that have multiple global minimums. Moreover, sensitivity analysis of the DE parameter shows that the parameter F, CR, NP, and G influences the optimal global function solution and computation time.

    Baca Full Text klik disini
    [thumbnail of S_MAT_1906080_Title.pdf] Text
    S_MAT_1906080_Title.pdf
    Download (242kB)
    [thumbnail of S_MAT_1906080_Chapter1.pdf] Text
    S_MAT_1906080_Chapter1.pdf
    Download (163kB)
    [thumbnail of S_MAT_1906080_Chapter2.pdf] Text
    S_MAT_1906080_Chapter2.pdf
    Restricted to Staf Perpustakaan
    Download (280kB) | Request a copy
    [thumbnail of S_MAT_1906080_Chapter3.pdf] Text
    S_MAT_1906080_Chapter3.pdf
    Download (398kB)
    [thumbnail of S_MAT_1906080_Chapter4.pdf] Text
    S_MAT_1906080_Chapter4.pdf
    Restricted to Staf Perpustakaan
    Download (1MB) | Request a copy
    [thumbnail of S_MAT_1906080_Chapter5.pdf] Text
    S_MAT_1906080_Chapter5.pdf
    Download (168kB)
    Official URL: http://repository.upi.edu/
    Item Type: Thesis (S1)
    Additional Information: ID Sinta Dosen Pembimbing: 1. 258640 KHUSNUL NOVIANINGSIH 2. 5981275 FITRIANI AGUSTINA
    Uncontrolled Keywords: Algoritma, Differential Evolution, optimal global, fungsi differentiable, fungsi non-differentiable.
    Subjects: L Education > L Education (General)
    Q Science > QA Mathematics
    Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Matematika - S1 > Program Studi Matematika (non kependidikan)
    Depositing User: Derry Romeo
    Date Deposited: 06 Feb 2023 04:33
    Last Modified: 06 Feb 2023 04:33
    URI: http://repository.upi.edu/id/eprint/87853

    Actions (login required)

    View Item View Item